The detection of criminal groups in real-world fused data: using the graph-mining algorithm \u201cGraphExtract\u201d

Chris Scogings,David Robinson

doi:10.1186/s13388-018-0031-9

Abstract

Law enforcement and intelligence agencies generally have access to a number of rich data sources, both structured and unstructured, and with the advent of high performing entity resolution it is now possible to fuse multiple heterogeneous datasets into an explicit generic data representation. But once this is achieved how should agencies go about attempting to exploit this data by proactively identifying criminal events and the actors responsible? The authors will outline an effective generic method that; computationally extracts minimally overlapping contextual subgraphs, then uses these subgraphs as the basis to construct a mesoscopic graph based on the intersections between the subgraphs, enabling knowledge discovery from these data representations for the purpose of maximally disrupting terrorism, organised crime and the broader criminal network.

Highlights

Law enforcement and intelligence agencies have relied on reactive sources of crime detection, such as the receipt of suspicious transactions, ‘tip offs’ from a covert human intelligence source (CHIS), or some other significant event
A subgraph that includes organised crime entities, a domestic corporate structure, a non-transparent offshore corporate structure, a series of assets owned by related parties, and a series of suspicious transaction reports can give an indication that a constellation of elements are present that represent the illicit generation, laundering, and realisation of proceeds
There is a prominent pattern of brokerage [44] with subgraph nodes 21302, 20848 and 5164 prevalent, as indicated by node size. Targeting these subgraphs could potentially yield a more enduring impact impairing the efficient functioning of the entire complex system. These findings indicate that across the entire criminal network, represented by the approximate 20,000 subgraphs within the mesoscopic graph, there is an outer periphery of 18,000 subgraphs indicating criminal activity largely unconnected to organised crime entities

Summary

Introduction

Law enforcement and intelligence agencies have relied on reactive sources of crime detection, such as the receipt of suspicious transactions, ‘tip offs’ from a covert human intelligence source (CHIS), or some other significant event (e.g. border interaction, search warrant, etc.). This paper outlines a novel graph mining method—the “GraphExtract” algorithm—that detects overlapping subgraphs (i.e. a collection of nodes and edges) of entities involved in atomic criminal events and generates a contextual view of how these subgraphs are connected to one another.

Results

Conclusion